SciELO - Scientific Electronic Library Online

Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Journal of the Chilean Chemical Society

versión On-line ISSN 0717-9707

J. Chil. Chem. Soc. v.48 n.2 Concepción jun. 2003 

J. Chil. Chem. Soc., 48, N 2 (2003)


Juan Ortiz, Marta Puelma, Juan Luis Gautier*

Laboratorio de Electroquímica. Departamento de Química de los Materiales. Facultad de Química y Biología.
Universidad de Santiago de Chile. Av. L.B.O'Higgins 3363 Santiago, Chile.

( Received : September 2, 2002 - Accepted : March 13, 2003)


We studied the oxidation of phenol by HO2- ions obtained from the oxygen reduction reaction (ORR) on pyrolytic graphite and on Ni0.3Co2.7 O4 spinel oxide in 1M KOH at 25C. Phenol degradation was achieved by using the controlled potential electrolysis of saturated O2 solutions under batch and laminar flow conditions. In both electrodes, the electro- generation of HO2- occurs between -0.2 V and -0.4V, with the graphite being more efficient than the NiO.3CO2.7 O4 electrode. The phenol degradation conversions for the batch electrolysis were 32% on graphite and 21% on the oxide, whereas with the flow mode they were 7% and 22% respectively.

Key Words: Peroxide ion, Oxidation, Phenol, Graphite, Nickel-Cobalt Oxide, Spinel Thin Films.


In the treatment of industrial wastewaters, biological agents do not efficiently eliminate phenol and the preferred option is the use of chemical oxidation [1]. In this regard, common oxidants that have been investigated include ozone, chlorine and hydrogen peroxide, the latter in presence of Fe2+ ions (Fenton's reagent) [2]. Hydrogen peroxide is advantageous because besides its high oxidative power (E= 1.77 V) it leads to less toxic, biodegradable products.

Hydrogen peroxide can be generated from the electroreduction of O2 either in the form of H2O2 in acidic medium or as HO2- in alkaline medium using adequate cathodic materials. It is well known that the oxygen reduction reaction (ORR) on graphite or glassy carbon cathodes leads mainly to H2O2. In this regard, there are several procedures related to the electro generation of H2O2 for direct or indirect organic oxidation that make use of carbon [3, 4], modified carbon [5], or a combination with Fenton's reagent [6-8]. On the other hand, mixed oxides with spinel structure have shown good electrocatalytic properties for the ORR in alkaline medium with formation of HO2- and OH - ions. This is the case of the NixCo3-xO4 system [9]. One of our recent studies with the NixCo3-xO 4 (0.1 £ x £ 1) system in alkaline solutions has shown that it is possible to indirectly oxidize ethylene glycol in high yields using the peroxide anions HO2- resulting from the cathodic reduction of O2 on Ni0.3Co2.7O4 [10]. This observation is in keeping with the limiting nickel concentration in the oxide (x < 0.6) necessary to obtain high electrocatalytical activity in the HO2- production as determined previously studying an oxide with extended composition, NixCo3-xO4 (0 £ x £ 1.8) [11]. The Ni0.3Co2.7O4 oxide with the cationic structure, Co2+0.43 Co3+0.57 [Co3+1.43 Co2+0.27 Ni3+0.3] shows the highest Co3+/Co2+ ratio in their structure as determined by XPS studies, which is very important since the Co3+/Co2+ redox couple has been considered essential for the electrogeneration of HO2- ions [10].

Phenol can also be destroyed directly by anodic oxidation in acidic solutions on Pt [12-14], Ta/PbO2 [15], Ti/IrO2, Ti/RuO2, Ti/SnO2, Ti/PbO2 [16,17] and Au [18] electrodes. By means of both a direct anodic oxidation and a chemical oxidation, the OH radicals that are generated destroy organic materials directly on the electrode surface by a mechanism that involves an electrophilic attack by OH on the organic compound and their oxidation products [12,15]. The initial products formed are hydroquinone, catechol and benzoquinone, which are further oxidized to maleic, fumaric and oxalic acids [19]. It is known that during the electrochemical oxidation of phenols a passive film is formed on the electrode surface which might be the result of the formation of tars coming from the electro polymerization of phenol [12-18,20,21] and which leads to slow reaction rates.

The double channel electrode flow cell (DCEFC) has been largely used in our laboratories to study the electrocatalytical behavior of mixed valence oxides [9,11] and sulfurs [22,23]. This technique allows working under diffusion-controlled conditions such as those in the RRDE technique.

The aim of this work was to evaluate the efficiency of a new method for phenol degradation. The method, based on the electro-generation of HO2- ions from O2 reduction both on graphite and on the spinel oxide Ni0.3Co2.7O4 is part of a continuous effort to use new materials for wastewater treatments under mild conditions.


1mol dm-3 KOH aqueous solutions saturated in O2 and containing 1000 ppm of phenol, were electrolyzed at constant potential (E) and a room temperature in both batch and laminar flows. Two types of cathodes were used: graphite and Ni0.3Co2.7O4 oxide. Films of the Ni0.3Co2.7O4 oxide (approximately 25 mm thickness) were deposited on Ti substrates by the spray pyrolysis technique [9-11]. Cobalt nitrate solutions were deposited only onto one side of the titanium foil by thermal decomposition of aqueous solutions of Ni(NO3)2 6H2O (Merck p.a. ref. 6721) and Co(NO3)2 6H2O (Fluka p.a. ref. 60833), mixed in the appropriate proportions. In order to avoid metal hydrolysis, 1L of the precursor solution was acidified with 1 mL of concentrated HNO3, and this solution was sprayed on titanium rectangular plates (Ti 07859 USA). The temperature of the Ti surface was continuously checked with a calibrated infrared pyrometer (Minolta Cyclops 330). The conditions of the spray technique are the same already reported [10], namely substrate at 350oC, air as vector gas at 2 atm, a solution flow rate of 2.5 cm3 min-1, 30 min of spraying time with the solution and 5 min of spraying time with water. The oxide film deposited on titanium was dried in air and finally annealed at 350C for 6 h. The cationic stoichiometry was determined by AAS (Perkin Elmer 2380), at wavelengths of 232 nm and 240.7 nm for Ni and Co, respectively. Phase purity was checked by X-ray diffractometry (XRD). The XRD analyses were carried out with a SIEMENS D-5000 diffractometer, using the CuKa radiation (l = 0.154056 nm). Surface morphology was controlled using a Jeol (JSM 840) SEM apparatus.

For the batch studies, a glass H-type cell was used in which the anodic and the cathodic compartments were separated by a sintered glass frit (Pyrex 4G). One compartment contained the cathode together with the reference electrode placed inside a tube with a Luggin capillary, and an inlet for the gases (N2 or O2). The other compartment contained a Pt wire having a large surface area. The cyclic voltammetry and the coulometric measurements were carried out using a Voltalab 40TM potentiostat connected to a Pentium PC.

The cathode of pyrolytic graphite (Carbone Lorraine, France) was a 1-cm2 disk exposed to electrolyte and the oxide electrode was a 0.8 cm2 (8 mm x 10 mm) rectangular plate. Both electrodes were connected to the external circuit using Cu wires.

In the case of the flow experiments, we used a double channel electrode flow cell (DCEFC) made out of acrylic, which has been described elsewhere [9]. The channel was 30 mm wide, 100 mm long and 1 mm deep. The generator electrode was 9 x 10 mm, and the collector electrode (Pt) was 10 x 15 mm. Both electrodes were fixed to the acrylic holder with epoxy resin (Araldite, Ciba) and mounted into the upper channel wall. The counter electrode (Pt) was embedded in the lower channel wall. The electrolyte was charged into the cell using a peristaltic pump (MasterFlex N 7521-00, 6-600 rpm) and the flow was controlled with a Cole-Palmer flow meter, with a 102.05 tube and a saphire float. It was possible to vary the flow in the range from 3,1mL min-1 to 126.3 mL min-1. The DCEFC was connected to a Pine RDE2 bipotentiostat and to two Graphtec WX 2300 X-Y recorders, which allowed for monitoring the potentials and measuring the currents of both the generator and the collector electrodes. All potentials were measured against an Hg/HgO/1M KOH reference electrode. The cell was calibrated using a potassium hexacyanoferrate (III) solution in deoxygenated KOH media. This calibration is useful to determine the collection efficiency factor N. The ORR was studied in a 1M KOH solution saturated with pure O2. The collector potential, Ec, was maintained at +0.7V, a value at which the oxidation of HO2- ions were diffusion-controlled. The actual HO2- ion concentration was evaluated by titration with acidic 0.05 M KMnO4 solution. The phenol analyses were performed in a Hitachi 655-12 HPLC apparatus with a L-5000 LC controller, a RP-18 (124 x 4) column, with acetonitrile-water (15: 85) as the mobile phase and UV detection at 270 mn. The aqueous phase was acidified with 1% H3PO4 solution. The injected sample volume was 20 mL, the eluent flow was 1mL min-1 and the pump pressure was fixed at 80 bar.

3. Results and discussion.

3.1 Cyclic Voltammetry measurements (CV).

3.1.1 Ni0.3Co2.7O4 electrode.

The oxide prepared using the spray pyrolysis technique showed a pure spinel phase Fd3m-type, with crystallites 35 nm in size and a unit a-cell parameter that measured 0.8085 nm in agreement with our previous reports [10].

Fig. 1 shows the oxide voltammograms in KOH solution in the absence of O2 (curve 1) and with 50 ppm of phenol (curve 2), and in the presence of O2 (curve 3) and with O2 and 50 ppm of phenol (curve 4). The scan was obtained from 0 V to -0.7 V. The presence of phenol increases the cathodic current and enlarges the voltammogram (curve 2) suggesting that phenol is adsorbed on the electrode surface. The O2 dissolved in the electrolyte starts its reduction at ­0.1 V according to the reaction O2 + H2O + 2e ®HO2- + OH- (curve 3). The presence of HO2- ions was confirmed by the formation of I2 upon using an acidic KI solution and soluble starch. Regarding curve 4 at -0.45 V, the I-E slope changes (in the cathodic branch) due to HO2- reduction according to the reaction HO2- + H2O + 2e ®3OH- . A large potential zone is not affected by the presence of phenol because its adsorption on the surface sites in this solution is less than that in a solution without O2, showing that both entities compete for the adsorption sites (see curves 2 and 4). However, starting at - 0.4 V the cathodic current increases more than the current observed in curve 3 (without phenol) indicating that phenol is preferentially adsorbed and limits the ORR that produces HO2-.

Fig.1 Cyclic Voltammograms of the Ni 0.3Co2.7O4 electrode in 1M KOH solution: (1) N2, saturated; (2) N2 saturated + 50 ppm of phenol; (3) O2 saturated; (4) O2 saturated + 50 ppm of phenol. Scan rate: 10 mVs-1.

3.1.2 Graphite electrode.

Fig. 2 shows the behavior of the graphite electrode in KOH saturated with O2 (curve 1) and in an O2 saturated solution plus 50 ppm of phenol (curve 2). The current peak observed at -0.3 V corresponds to the O2 reduction that gives rise to HO2- ions. The scan rate for these curves was 100 mV s-1 indicating that the ORR is faster than on the oxide electrode. The current peak decreases to 50% when 50 ppm of phenol were added to the solution which could be a result adsorption over the electrode surface, either by phenol or by its oxidation products, a process already suggested in similar situations [20, 21].

Fig. 2 Cyclic Voltammorams of the graphite electrode of in KOH 1M solution: (1) O2 saturated; (2) O2 saturated + 50 ppm of phenol. Scan rate: 100 mVs-1.

  3.2 Studies using the double channel electrode flow cell (DCEFC).

3.2.1 Theoretical review on ORR

At the cathodically polarized generator electrode, the dissolved oxygen in the electrolyte is reduced directly to OH- ions by a 4-electron transfer pathway (reaction 1, rate constant k1) and concurrently to the hydrogen peroxide ions, HO2-, as shown in the indirect 2-electron pathway (reaction 2, rate constant k2). After this, the peroxide ion reduction to OH- ions takes place (reaction 3, rate constant k3) which gives rise to the well-known 2+2 mechanism:

Direct oxygen reduction to OH- ions, 4e-pathway (k1):
O2 + 2H2O + 4e ®4 OH-

Oxygen reduction to HO2- ions, 2e-pathway (k2):
O2 + H2O + 2e ®HO2- + OH-


Peroxide ion reduction to OH- ions, 2e-pathway (k3):
HO2- + H2O + 2e ®3 OH-


The HO2- ions produced at the generator electrode are transported by the electrolyte flow to the collector electrode where they are oxidized to O2 and generate Ic (current at the collector electrode) according to the collection efficiency factor N. Also, it is possible that the HO2- ions undergo chemical disproportonation to O2 and OH- ions. Both the direct (k1) and the indirect reactions (k2 and k3) generate Ig (current at the generator electrode).

The DCEFC method allows us to determine the best conditions (potential, flow, etc.) for the preparation of the HO2- ions, on the basis of the evaluation of the kinetic constant ratio (k1/k2), and ng, the number of electrons exchanged in the ORR. Equations (4) and (5) can be obtained applying the steady state approximation [9]:

where v is the linear electrolyte flow rate, and B and C are constants. Equations (4) and (5) correspond to linear variations that can be represented in a Cartesian coordinate plot. From the intercept of the ­N(Ig/Ic) vs. v-1/3 plot, one can obtain the ratio k1/k2 which indicates the relative contributions of the direct and parallel reduction pathways. Then, if k1/k2< 1 the peroxide ions formation is favored, whereas if k1/k2 > 1 the OH- ions formation is favored. The slope of the -1/Ig vs v-1/3 plot allows determining the experimental value ng at each potential.

3.2.2 O2 reduction on the Ni0.3Co2.7 O4 and on graphite electrodes

From the Ig and Ic measurements vs. v-1/3 (plots not shown) it was possible to suggest the following ORR (6) for both electrodes:

This reaction is the consequence of the reactions 6a and 6b:

Indeed, within the potential range explored (from -0.6V to -0.1V), the exchanged electron number, ng, varies from 3 to 2 independently of the chemical nature of the generator electrode (Fig. 3). On the Ni-Co oxide the ORR occurs at a lower potential than on graphite, which shows the electrocatalytical importance of the mixed oxide in this reaction. On the other hand, reaction 6b becomes important starting at E £ - 0.25 V for the oxide electrode and at E £ - 0.4 V for the graphite electrode, as evidenced by the k1/k2 > 1 value (Fig.4). Therefore, the best conditions for the production of HO2- ions consist of E values higher than -0.25 V for the oxide electrode and higher than -0.4 V for the graphite electrode.

Fig. 3 Variation of the number of exchanged electrons (ng) with the applied potential in the ORR (1) on the Ni 0.3Co2.7O4 electrode and (2) on the graphite electrode in 1M KOH solution saturated with O2. Fig. 4 Variation of the k1/ k2 ratio with the applied potential in the ORR using (1) the Ni 0.3Co2.7O4 electrode and (2) the graphite electrode.

3.3 Production efficiency of HO2- at controlled potential.

On the basis of our previous results with CV and DCEFC, the chosen potentials were -0.35 V and -0.15 V for the graphite and the oxide electrodes, respectively. At these values only the reaction 6a occurs and it corresponds to a k1/k2 ratio of 0.7 on both electrodes. We carried out electrolysis at constant potential with several duration times. At the end of each experience samples of the electrolyte were extracted in order to determine their HO2- ion content. We also measured the electric charge consumed at various times in the interval 30 to 120 min of each electrolysis. The HO2- production efficiency (h), was calculated using the equation (7):

h = 2Fn1007Q


where F is the Faraday constant, n the number of moles of HO2- ions in the solution, obtained by titration with aqueous 0.05 M KMnO4 and, Q is the electric charge measured at time t.

Fig. 5 shows the efficiency of the generation of HO2- ions in the oxygen reduction as a function of the electrolysis time. It is interesting to observe that after 1 hour the graphite electrode shows an almost constant efficiency with time, which on the average is about nine times higher than the one observed with the oxide electrode. In the case of the oxide electrode, the efficiency decreases with time. This behavior may be related with the different O2 adsorption mechanisms that can take place on the surface of both electrodes.

Fig. 5 Efficiency of generation of HO2- ion as a function of the electrolysis time at constant E. (1) Ni 0.3Co2.7O4 electrode at -0.15 V; (2) graphite electrode at -0.35 V.

3.4 Phenol oxidation efficiency

Using the graphite and oxide electrodes we carried out electrolysis at constant potential for 2 hours on KOH solutions that contained about 1000 ppm of phenol. The phenol content was measured at regular time intervals. The electrolysis was carried out using the electrolyte both in batch system and in continuous flow modes. Figure 6 shows the decay in phenol concentration for the two electrodes in the batch mode. With both electrodes, at short electrolysis time (about 20 min) the concentration of phenol decreases to reach a constant value: 650 ppm in graphite and 750 ppm in the oxide. It is clear that the phenol oxidation products are adsorbed on the surfaces of both types of electrode producing a blocking effect for additional O2 reduction.

Fig. 6 Oxidation of phenol at constant E as a function of the electrolysis time in the presence of O2 and 1000 ppm of phenol: (1) Ni 0.3Co2.7O4 electrode at -0.15 V; (2) graphite electrode at -0.35 V . Batch system.

The efficiency in phenol oxidation is expressed as percent conversion, %ox = (Ci - Ct)/Ci)100, where Ct is the phenol concentration at time t and Ci the initial phenol concentration. After 2 h of electrolysis with the graphite electrode the conversion was 32% whilst with the Ni0.3Co2.7O4 electrode was only 21%.

The inverse effect was observed when using the DCEFC system at 29 mL min-1 of electrolyte flow, which is shown on Fig 7. Indeed, the concentration of phenol when the graphite electrode is used reaches 900 ppm after 20 minutes of electrolysis (7% conversion) whilst with the oxide electrode is 750 ppm (22% conversion) in the same period of time. Since the oxidation efficiency on the graphite electrode diminishes when the electrolyte flows, it is clear that the electrolysis conditions determine the indirect phenol oxidation.

Fig.7 Oxidation of phenol in DCEFC as a function of the electrolysis time in presence of O2 and 1000 ppm of the phenol at constant E. (1) Ni0.3Co2.7O4 electrode at - 0.15 v (") graphite electrode at - 0.35 v. Electrolyte flow: 29 mLmin-1.

Despite the fact that the oxide produces smaller yields of phenol degradation when compared with using Fenton's reagent, it presents several advantages in comparison to the graphite electrode: low electrode potential, high effectiveness using flow conditions and easy electrode preparation in thin film form. Unfortunately the oxide electrodes cannot be used in acidic medium where the Fenton-based yield is maximum because an important corrosion takes place. However it is possible to prevent this corrosion by using a protective polymer thin film such as polypyrrole (PPy) which forms a multilayer composite electrode: GC/PPy/PPy(Ox)/PPy [24]. This composite electrode with nanoparticles of Ox = Ni0.3Co2.7O4 can also catalyze the ORR as it has been recently demonstrated [25]. Further research is necessary using the oxide-polymer composite electrode to achieve high phenol conversion in acidic media.


The generation of HO2- in alkaline medium takes place in the (-0.2 V) ­ (-0.4 V) potential range with both electrodes. On the graphite electrode the HO2- generation efficiency is around 9 times greater than that on the Ni0.3Co2.7O4 spinel electrode. The phenol degradation at controlled E depends on the type of system used (batch vs. flow), which can induce a local HO2- variation. Using the batch system, 32% and 21% of phenol are oxidized by the HO2- ions electro generated on the graphite electrode (at -0.35 V) and on the oxide electrode (at -0.15 V), respectively. In the case of the flow electrolyte system (DCEFC), the phenol conversion reached 7% at -0.35 V on the graphite electrode and 22% on the oxide electrode at -0.15 V. In both systems, the oxide produces similar amounts of HO2-.

Our results show that the oxidation of phenol also leads to polymeric products on both types of electrodes, which affects their performance. We consider that it is feasible to increase the indirect phenol degradation by using HO2- electro generated on a Ni0.3Co2.7O4 electrode in DCEFC upon adding Fenton's reagent, but for this to happen it would be necessary to protect the electrode against corrosion by acid. Despite this, the use of the oxide electrode presents several advantages, including a low potential for the electrolysis, better efficiency under flow conditions, and a very small mass because it can be used in thin films.


Financial support from DICYT-USACH, CNRS-CONICYT and FONDECYT (Project 1020066) are gratefully acknowledged.


1. J.-S.Do and W.C.Yeh. J.Appl.Electrochem. 26 (1996) 673         [ Links ]

2. N. Al-Hayek, M. Dore, Environmental Technology lett. 6 (1985) 37         [ Links ]

3. P.C. Foller and R.T.Bombard. J.Appl. Electrochem. 25 (1995) 613         [ Links ]

4. C.Ponce de León, D.Pletcher.J.Appl.Electrochem. 25 (1995) 307         [ Links ]

5. J-S. Do, C.- P.Chen, J. Appl.Electrochem. 24 (1994) 936         [ Links ]

6. M.Sudoh, T.Kodera, K.Sakai, J.Q.Zhang, K.Koide, J.Chem. Engin. Japan, 19 (1986) 513         [ Links ]

7. A.Alvarez-Gallegos and D.Pletcher. Electrochim. Acta 44 (1998) 853         [ Links ]

8. A.Alvarez-Gallegos and D.Pletcher.Electrochem.Acta 44 (1999) 2483         [ Links ]

9. N.Heller-Ling, G.Poillerat, J.F.Koenig, J.L.Gautier, P.Chartier, Electrochim. Acta 39 (1994) 1669         [ Links ]

10. E.Ríos, H.Nguyen-Cong, J.F.Marco, J.R.Gancedo, P.Chartier, J.L.Gautier Electrochim. Acta 25 (2000) 4431         [ Links ]

11. N. Heller-Ling, M. Prestat, J.L. Gautier, J.- F. Koenig, G. Poillerat, P. Chartier, Electrochim. Acta 42 (1997) 197         [ Links ]

12. Ch.Comninellis, C.Pulgarin J. Appl. Electrochem. 21 (1991) 703         [ Links ]

13. M.Gatrell, D.W.Kirk J. Electrochem. Soc. 140 (1993) 903         [ Links ]

14. M.Gatrell, D.W.Kirk J. Electrochem. Soc. 140 (1993) 1534         [ Links ]

15. N.B.Tahar, A.Savall J. Electrochem. Soc. 145 (1998) 3427         [ Links ]

16. Ch. Comninellis, Trans I Chem. Engineering 70 B (1992) 219         [ Links ]

17. Ch. Comninellis, A.Nerini. J. Appl. Electrochem. 25 (1995) 23         [ Links ]

18. J.Wang, M.Jiang, F.Lu. J. Electroanal. Chem. 444 (1998) 127         [ Links ]

19. Ch. Comninellis , Trans IChemE, 70, Part B, (1992) 219         [ Links ]

20. A.Courteix, A.Bergel, M.Contat. J. Appl. Electrochem. 25 (1995) 508         [ Links ]

21. M.S.Ureta-Zañartu, P.Bustos, M.C.Diez, M.L.Mora, C.Gutierrez, Electrochem. Acta 46 (2001) 2545         [ Links ]

22. J.L.Gautier, J.Ortiz, Heller Ling, G.Poillerat, P.Chartier, J.Appl. Electrochem. 28 (1998) 827         [ Links ]

23. J.Ortiz, S.Barbato, J.L.Gautier, Bol. Soc. Chil. Quím., 45 (2000) 441         [ Links ]

24. H.Nguyen-Cong , El Abassi, P.Chartier Electrochem. and Sol. Stat. Lett. 3 (2000) 192         [ Links ]

25. H.Nguyen-Cong, V. de la Garza Guadarrama, J.L.Gautier, P.Chartier, J. New Mat. Electrochem. Systems 5 (2002) 35         [ Links ]

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons